In this article, an optimal bipartite consensus control (OBCC) scheme is proposed for heterogeneous multiagent systems (MASs) with input delay by reinforcement learning (RL) algorithm. A directed signed graph is established to construct MASs with competitive and cooperative relationships, and model reduction method is developed to tackle input delay problem. Then, based on the Hamilton–Jacobi–Bellman (HJB) equation, policy iteration method is utilized to design the bipartite consensus controller, which consists of value function and optimal controller. Further, a distributed event‐triggered function is proposed to increase control efficiency, which only requires information from its own agent and neighboring agents. Based on the input‐to‐state stability (ISS) function and Lyapunov function, sufficient conditions for the stability of MASs can be derived. Apart from that, RL algorithm is employed to solve the event‐triggered OBCC problem in MASs, where critic neural networks (NNs) and actor NNs estimate value function and control policy, respectively. Finally, simulation results are given to validate the feasibility and efficiency of the proposed algorithm.

We propose a contrastive dimensionality reduction approach (CDR) for interactive visual cluster analysis. Although dimensionality reduction of high-dimensional data is widely used in visual cluster analysis in conjunction with scatterplots, there are several limitations on effective visual cluster analysis. First, it is non-trivial for an embedding to present clear visual cluster separation when keeping neighborhood structures. Second, as cluster analysis is a subjective task, user steering is required. However, it is also non-trivial to enable interactions in dimensionality reduction. To tackle these problems, we introduce contrastive learning into dimensionality reduction for high-quality embedding. We then redefine the gradient of the loss function to the negative pairs to enhance the visual cluster separation of embedding results. Based on the contrastive learning scheme, we employ link-based interactions to steer embeddings. After that, we implement a prototype visual interface that integrates the proposed algorithms and a set of visualizations. Quantitative experiments demonstrate that CDR outperforms existing techniques in terms of preserving correct neighborhood structures and improving visual cluster separation. The ablation experiment demonstrates the effectiveness of gradient redefinition. The user study verifies that CDR outperforms t-SNE and UMAP in the task of cluster identification. We also showcase two use cases on real-world datasets to present the effectiveness of link-based interactions.

Deep learning based subspace clustering methods have attracted increasing attention in recent years, where a basic theme is to non-linearly map data into a latent space, and then uncover subspace structures based upon the data self-expressiveness property. However, almost all existing deep subspace clustering methods only rely on target domain data, and always resort to shallow neural networks for modeling data, leaving huge room to design more effective representation learning mechanisms tailored for subspace clustering. In this paper, we propose a novel subspace clustering framework through learning precise sample representations. In contrast to previous approaches, the proposed method aims to leverage external data through constructing lots of relevant tasks to guide the training of the encoder, motivated by the idea of meta-learning. Considering limited layer structures of current deep subspace clustering models, we intend to distill knowledge from a deeper network trained on the external data, and transfer it into the shallower model. To reach the above two goals, we propose a new loss function to realize them in a joint framework. Moreover, we propose to construct a new pretext task for self-supervised training of the model, such that the representation ability of the model can be further improved. Extensive experiments are performed on four publicly available datasets, and experimental results clearly demonstrate the efficacy of our method, compared to state-of-the-art methods.

Dimensionality Reduction (DR) techniques can generate 2D projections and enable visual exploration of cluster structures of high-dimensional datasets. However, different DR techniques would yield various patterns, which significantly affect the performance of visual cluster analysis tasks. We present the results of a user study that investigates the influence of different DR techniques on visual cluster analysis. Our study focuses on the most concerned property types, namely the linearity and locality, and evaluates twelve representative DR techniques that cover the concerned properties. Four controlled experiments were conducted to evaluate how the DR techniques facilitate the tasks of 1) cluster identification, 2) membership identification, 3) distance comparison, and 4) density comparison, respectively. We also evaluated users' subjective preference of the DR techniques regarding the quality of projected clusters. The results show that: 1) Non-linear and Local techniques are preferred in cluster identification and membership identification; 2) Linear techniques perform better than non-linear techniques in density comparison; 3) UMAP (Uniform Manifold Approximation and Projection) and t-SNE (t-Distributed Stochastic Neighbor Embedding) perform the best in cluster identification and membership identification; 4) NMF (Nonnegative Matrix Factorization) has competitive performance in distance comparison; 5) t-SNLE (t-Distributed Stochastic Neighbor Linear Embedding) has competitive performance in density comparison.

Nowadays, as data becomes increasingly complex and distributed, data analyses often involve several related datasets that are stored on different servers and probably owned by different stakeholders. While there is an emerging need to provide these stakeholders with a full picture of their data under a global context, conventional visual analytical methods, such as dimensionality reduction, could expose data privacy when multi-party datasets are fused into a single site to build point-level relationships. In this paper, we reformulate the conventional t-SNE method from the single-site mode into a secure distributed infrastructure. We present a secure multi-party scheme for joint t-SNE computation, which can minimize the risk of data leakage. Aggregated visualization can be optionally employed to hide disclosure of point-level relationships. We build a prototype system based on our method, SMAP, to support the organization, computation, and exploration of secure joint embedding. We demonstrate the effectiveness of our approach with three case studies, one of which is based on the deployment of our system in real-world applications.

Early active learning, in a common paradigm, usually selects representative samples for human annotating. This aligns with the goal of minimizing the overall reconstruction error in an unsupervised manner. While existing methods mainly focus on data samples that are drawn from individual yet high-dimensional feature space, they can hardly handle the real-world scenario where samples are often represented by low-dimensional features drawn from multiple groups (subspaces). In this case, how to leverage the grouping structure to select most representative samples becomes the key point to success. In this paper, we propose an unsupervised active learning framework, called Robust Grouping Active Learning (RGAL), to achieve this goal. The key idea is to take into account of different degrees of information shared across data groups. Specifically in RGAL, we assume data from some group can be embedded in a low-dimensional space, as well as that the data distributions of different groups can overlap with each other to a certain degree. And RGAL controls such group overlaps by imposing sparsity constraints on a matrix of reconstruction coefficients. To encourage a smooth coefficient space, we also enforce a robust loss with Laplacian regularization for noise suppression. We perform extensive experiments on multiple tasks which normally require costly human annotation, including facial age estimation, video action recognition and medical image classification. Results on benchmark datasets clearly demonstrate the efficacy of our RGAL method compared state-of-the-art methods.

The mini-challenge 2 of VAST Challenge 2019 asks the participants to make sense of the radiation conditions in St. Himark using radiation readings from the both stationary monitors and mobile sensors, particularly, to detect and monitor a bunch of contaminated cars running in the city. This paper presents our visual analysis solution to detect and localize these cars using various visualization techniques, including small multiples, distribution histogram, and animation. As a result, we detected most of these cars, characterize their behavior during the given time period and give suspected locations of these cars at the end. We also developed a visual analysis interface using Tableau to help users evaluate the uncertainty of sensor readings, make sense of radiation changes in different area, and make future plans to deploy more sensors.

In this paper, we develop a visual analysis method for interactively improving the quality of labeled data, which is essential to the success of supervised and semi-supervised learning. The quality improvement is achieved through the use of user-selected trusted items. We employ a bi-level optimization model to accurately match the labels of the trusted items and to minimize the training loss. Based on this model, a scalable data correction algorithm is developed to handle tens of thousands of labeled data efficiently. The selection of the trusted items is facilitated by an incremental tSNE with improved computational efficiency and layout stability to ensure a smooth transition between different levels. We evaluated our method on real-world datasets through quantitative evaluation and case studies, and the results were generally favorable.

In this paper, we mainly study the method of estimating the opponent's behavior in the confrontation environment.Firstly, the method of estimating the motion state of the other agent is studied.Then, the state estimation algorithm is designed, and weighted average filter is used to improve the state estimation of moving targets.Finally, the effectiveness of the algorithm is proved by experiments.Emphasizes the importance of intelligent learning.

This paper presents a methodology to calculate Voltage Distribution Factors (V-DFAX) for post-contingency low-voltage violations. For low-voltage violations which have no controlling action besides post-contingency load-shedding, PJM and their Transmission Owerns (TOs) develop Post-Contingency Local Load Relief Warning (PCCLRW) plans which are only implmented in the unlikely event the contingency occurs in real-time operations. Load shedding Distribution Factors (DFAX) for thermal violations are calculated within the Energy Management System (EMS). However, due to the non-linear relationship between load and voltage, a gap exists for V-DFAX. The proposed methodology is to address this gap by using graph theory and AC power flow solutions. The methodology and its implementation are presented in the context of PJM's power grid.

This paper presents a new methodology to detect low-frequency oscillations in power grids by use of time-synchronized data from phasor measurement units (PMUs). Principal component analysis (PCA) is first applied to the massive PMU data to extract the low-dimensional features, i.e., the principal components (PCs). Then, based on persistent homology, a cyclicity response function is proposed to detect low-frequency oscillations through the use of PCs. Whenever the cyclicity response exceeds a numerically robust threshold, a low-frequency oscillation can be detected instantly. Such swift detection can then be followed by modal analysis tools for more detailed information about the oscillation. Numerical examples using real data illustrate the effectiveness of the proposed methodology for quick detection of oscillations during operations.

This paper investigates how to perform online system identification employing synchrophasor data. Two approaches to identifying a reduced-order model are presented: a purely data-driven approach, and an approach that integrates online data-driven dynamic system identification with first-principle offline selective modal analysis. With prior knowledge of the frequency range interesting to power system operators, it is shown that the second approach recovers the key modes of the original system and produces a much reduced-order model of grid-level dynamics. Even with the presence of uncertainty about the actual modes of interest, an automatic tuning scheme is devised to adaptively adjust the frequency range to improve system identification. Numerical examples with synthetic synchrophasor data demonstrate the efficacy of the proposed identification approach.

Tracking how correlated ideas flow within and across multiple social groups facilitates the understanding of the transfer of information, opinions, and thoughts on social media. In this paper, we present IdeaFlow, a visual analytics system for analyzing the lead-lag changes within and across pre-defined social groups regarding a specific set of correlated ideas, each of which is described by a set of words. To model idea flows accurately, we develop a random-walk-based correlation model and integrate it with Bayesian conditional cointegration and a tensor-based technique. To convey complex lead-lag relationships over time, IdeaFlow combines the strengths of a bubble tree, a flow map, and a timeline. In particular, we develop a Voronoi-treemap-based bubble tree to help users get an overview of a set of ideas quickly. A correlated-clustering-based layout algorithm is used to simultaneously generate multiple flow maps with less ambiguity. We also introduce a focus+context timeline to explore huge amounts of temporal data at different levels of time granularity. Quantitative evaluation and case studies demonstrate the accuracy and effectiveness of IdeaFlow.

Online discussion forums make up a significant bulk in the type of opinion information that represents a valuable source for many real-world applications. However, conducting comprehensive opinion analysis of threaded discussions is a challenging task because it requires not only an aggregation of opinions over the multi-level thread structures, but also effective methods for exploring the complex relationships across different aggregated levels. In this paper, we present a visual analysis approach to address this challenge. Our approach leverages efficient text analysis methods to extract opinions and topical structures from massive threaded discussion data, and provides integrated visualizations to convey both opinion and threaded discussion structures. A suite of interaction tools is provided to enable cross-level explorations of opinions. We demonstrate the effectiveness and efficiency of the approach by conducting a case study on a real-world threaded discussion data.

Data mining is able to automatically analyze massive data in the data reservoir to obtain rich warning knowledge,which can be applied in the crisis warning;therefore it plays an important role in the safety operation of power equipment. Considering the time complexity,this paper,first of all,mines the historic data of the power transformation equipment through the improved clustering algorithm and obtains the data reservoir of the equipment operation status,and then provides real-time prediction values of the power transmission equipment by the regression algorithm. The related warning rules are used to realize the early warning for the equipment monitoring and finally based on the power transmission equipment monitoring data mining an early warning mechanism design is proposed.

Low short circuit capacity and high grid impedance are the main electrical characteristics of weak grid, and there are some stability and power quality problems with a large number of PV power generation systems incorporated into weak grid. The increasing penetration of PV power generation easily lead to voltage rise at the end of grid, and it affects the ability of weak grid to accept PV power capacity. This paper analyzes the impact of the line impedance, X/R ratio and power factor of PV inverter on voltage deviation, proposes an adaptive control strategy to enhance penetration of PV power generation in weak grid. This method determines the optimal power factor of different grid-connected PV inverter according to the grid X/R values, and control the output active power and reactive power to improve PV penetration under weak grid condition. Simulation results demonstrate the effectiveness of the proposed method.

This paper explores a potential approach to fast classifying power system events using online synchrophasor measurements. The approach is based on dimensionality reduction of the emerging ambient phasor measurement unit (PMU) data. In contrast with model-based analysis, the proposed approach does not require a system model. It projects real-time PMU data onto the core subspace constructed from pre-event data, and then utilizes their scatter plots to detect and classify the system events. Projections lying outside the core subspace indicate the occurrence of an event, and the topological shapes of these projections classify the events. Numerical examples using synthetic PMU data are conducted to demonstrate the efficacy of the proposed approach.

This paper studies the fundamental dimensionality of synchrophasor data, and proposes an online application for early event detection using the reduced dimensionality. First, the dimensionality of the phasor measurement unit (PMU) data under both normal and abnormal conditions is analyzed. This suggests an extremely low underlying dimensionality despite the large number of the raw measurements. An early event detection algorithm based on the change of core subspaces of the PMU data at the occurrence of an event is proposed. Theoretical justification for the algorithm is provided using linear dynamical system theory. Numerical simulations using both synthetic and realistic PMU data are conducted to validate the proposed algorithm.

This paper reports our recent work on dimensionality reduction of synchrophasor data and subsequent engineering analysis of the results. Principal component analysis (PCA) based dimensionality reduction is first applied to explore the underlying dimensionality of power systems from the data of massively deployed PMUs. Then the physical interpretations are provided with the power engineering insights: spatial interpretation suggests the coherency of generator groups; temporal analysis indicates the time-scale hierarchy of power system operations. Numerical examples using both synthetic and realistic PMU data are conducted to illustrate the potential value of combining PMU data-driven and physics-based analytics in real-time grid operations.